Wealth Engine
Long Context Models
-
Tencent Hy4: 770B Open Weights, 82% Cheaper Output Than Kimi K3
Tencent Hy4 ships 770B open weights under Apache 2.0 with a 1M-token context and $2.501 output pricing, 82% below Kimi K3. Benchmarks, costs, catches.
-
Cheapest 1M Context Model: GLM-5.3-Flash vs Gemini 3.7 Flash
The cheapest 1M context model just changed. GLM-5.3-Flash costs 5x less than Gemini 3.7 Flash and scores higher on intelligence. Full breakdown.